(a) Discuss the signal detection theory (SDT) with reference to perceptual vigilance task performance. (20 marks) (b) Does…

(a) Perceptual vigilance and SDT

Signal detection theory treats perception as a decision under uncertainty. It separates sensitivity, d′, the separation between signal and noise distributions, from response bias, β, the criterion for saying “target.” ROC curves plot hit rate against false-alarm rate across criteria, showing that the same d′ can yield different performance if the observer is liberal or conservative. In practice, ROC analysis can reveal whether a vigilance decrement is mainly a loss of discrimination or a change in decision rule. In perceptual vigilance, targets are rare and monitoring is prolonged. Performance may decline because d′ falls—due to fatigue, monotony, low target probability, high workload, poor arousal or motivation—or because β shifts conservatively, reducing hits and false alarms. Both changes matter: demanding tasks can show genuine sensitivity decrements, while low-risk monitoring often shows criterion change. Mackworth’s clock test, where a pointer moving in steps round a clock face occasionally made a rare double jump, demonstrated a time-on-task decrement. Wiener’s vigilance studies extended this by showing that sustained attention depends on allocation of resources and response policy. Applied contexts such as radar monitoring, air traffic control and security screening show that SDT helps distinguish perceptual failure from institutional bias.

(b) Punishment: conditional utility

Punishment does not work in all situations. It can suppress a behaviour when it is contingent on the response, immediate, sufficiently intense, consistent, and paired with reinforcement of alternatives. Punishment is most defensible when mild, proportionate and temporary, and least effective when used in anger or as the only strategy. Its advantages are speed, safety, and deterrence when the punisher is predictable. Its disadvantages are fear, aggression, escape, avoidance, damage to trust, and failure to teach better behaviour. It must be distinguished from negative reinforcement, which removes an aversive stimulus to increase a behaviour, whereas punishment adds one to decrease it. Skinner treated punishment as powerful but risky, emphasising side-effects; Azrin-Holz argued it can be effective under strict conditions but is not universally appropriate. In India, NCPCR's national survey (2009-10) found that almost all sampled children reported some corporal punishment, and Section 17 of the RTE Act, 2009 now prohibits it, yet evidence links it to anxiety, aggression and disengagement rather than durable learning. Criminal-justice debates over deterrence, especially capital punishment, show that punishment may deter some offences but is weakened by low detection and conviction rates. In organisations, harsh discipline may reduce immediate violations but can increase resentment, turnover and concealment.

(c) Reconstructed memory

Bartlett’s War of the Ghosts showed that memory is reconstructive rather than reproductive. Instead of replaying a stored trace, people rebuild experiences using schemas described in his schema theory, expectations and current goals. Encoding is already selective and interpretive; retrieval is a generative act in which gaps are filled, details are omitted, and events are made coherent. Loftus’s misinformation paradigm demonstrated that post-event information can alter reports, especially when questions are leading. Eyewitness-testimony research shows that stress, delay, weapon focus, and social pressure reduce accuracy. Indian legal and trauma studies, including work on child witnesses and sexual-assault cases, similarly show that suggestibility, repeated questioning and emotional stress can distort recall. Thus reconstructed memory is not simply inaccurate; it is accurate to the extent that it serves present meaning, but it can be seriously misleading in forensic contexts.